| Abstract|| |
A good vascular access is the lifeline of patients on long-term hemodialysis (HD) and anteriovenous fistula is considered the ideal access. Vascular access related septicemia (VARS) is the second most common cause of mortality among HD patients. Such infections could also lead to loss of vascular access unless specific measures are taken to preserve the accesses. The present study was designed to determine the incidence of septicemia, common bacterial flora involved, and impact of early, empirical antibiotic therapy on vascular access salvage among HD patients. This prospective study, involved 209 patients, undergoing long-term HD, from June 1996 to June 2000. A total of 85 (40.6%) developed VARS with predominance in females (63.7%), patients above 50 years of age (37.0%) and those having diabetes mellitus (25.1%). A total of 124 episodes of septicemia were recorded with an average of 1.23 episodes per 100 patient-months during the four year (10032 patient-months) study period. Peripheral blood samples for culture and sensitivity were collected and the patients were started empirically on amikacin-vancomycin combination which was modified after obtaining culture and sensitivity results. A cure was defined as 45 days symptom-free interval after antibiotic therapy was completed. Staphylococcus aureus was the commonest (29.0%) organism associated with VARS, followed by Pseudomonas aeruginosa (15.3%). The temporary vascular access group recorded maximum number of VARS episodes; [femoral catheter (FC) group, (43.5%), followed by subclavian (SC) group, (28.2%)] and the lowest (8.8%) was seen in the AVF group. Vascular access salvage rate of 48/85 (56.4%) and mortality of 22/85 (25.9%) was observed in the present study. Antibiotic access salvage with AmikacinVancomycin combination has an advantage of preserving vascular access sites in at least, 50% of cases.
Keywords: Hemodialysis, Septicemia, Vascular access, Antibiotic salvage.
|How to cite this article:|
Saxena AK, Panhotra B R, Naguib M, Sundaram D S, Venkateshhappa C K, Uzzaman W, Al-Mulhim K. Vascular Access Related Septicemia in Hemodialysis: A Focus on Bacterial Flora and Antibiotic Access Salvage. Saudi J Kidney Dis Transpl 2002;13:29-34
|How to cite this URL:|
Saxena AK, Panhotra B R, Naguib M, Sundaram D S, Venkateshhappa C K, Uzzaman W, Al-Mulhim K. Vascular Access Related Septicemia in Hemodialysis: A Focus on Bacterial Flora and Antibiotic Access Salvage. Saudi J Kidney Dis Transpl [serial online] 2002 [cited 2020 Jan 26];13:29-34. Available from: http://www.sjkdt.org/text.asp?2002/13/1/29/33801
| Introduction|| |
Hemodialysis (HD) delivery requires safe and reliable access to blood vessels capable of providing rapid extracorporeal blood flow and primary arteriovenous fistulas (AVF) are the preferred form of vascular access.  The Dialysis Outcomes Quality Initiative (DOQI) guideline-working panel recommends having AVF in at least 50% of the prevalent end-stage renal disease (ESRD) population. However, improper vascular anatomy due to co-morbid conditions often preclude creation of AVF, mostly in elderly and diabetic patients. Late referral for vascular access and the indiscriminate use of veins for phlebotomy also reduce the likelihood of successful AVF creation.  In addition, AVF requires two to four months to mature and even with careful screening for native fistula placement, 30 to 40% fail to mature. Consequently, less desirable modes of vascular access like arteriovenous graft (AVG), permanent subcutaneous tunneled cuffed catheters (PC) and temporary vascular catheters have to be increasingly relied upon. In fact, up to 60% of new patients and 30% of prevalent patients on HD are using a catheter for dialysis access.  The life span of the vascular access is adversely affected by complications like thrombosis and infections.  Thus, septicemia due to access infection remains the most common cause of mortality, second only to cardiovascular deaths, among HD patients. , Bacterial colonization of access catheters and infection at the local site occur frequently either due to contamination of catheter lumen or, migration of bacteria from skin through the entry site, and remains the common source of septicemia in these patients. 
Staphylococcus aureus is the commonest bacteria associated with septicemia among these patients, followed by gram-negative bacilli. , Immunosuppression, old age and diabetes mellitus enhance the susceptibility to septicemic illness.  In addition, infection also leads to loss of vascular catheters and failure of AVF and AVG. The management of septicemia would naturally depend on the prevalent bacteria and the sensitivity pattern to antibiotics might vary in different centers.  However, Marr et al, have proposed antibiotic salvage of vascular catheters as an alternative to catheter removal in their prospective Duke University preliminary trial. 
The present study was undertaken to determine the incidence of septicemia, commonly associated bacteria, colonization of the vascular catheters, and to assess the role of an early empirical antibiotic therapy on vascular access salvage, among patients on long-term HD.
| Materials and Methods|| |
The present prospective study was undertaken on 209 patients with ESRD undergoing long-term HD at the King Fahad Hospital, Hofuf, Saudi Arabia, from June 1996 to June 2000. There were 96 males and 113 females. These patients were dialyzed two or three times a week with cuprophane membrane dialyzers through the following vascular accesses:
a) subclavian dual lumen catheter (SC),
b) femoral dual lumen catheter (FC),
c) permanent subcutaneous tunneled cuffed catheter (PC),
d) arteriovenous grafts (AVG) and
e) arteriovenous fistula (AVF).
There was no clinical evidence of septicemia at the time of creating these vascular accesses. The clinical criteria for diagnosis of septicemia was based on the presence of backache, fever, chills and rigors, unexplained nausea, vomiting and hypotension, mental changes and leukocytosis. 
Peripheral blood culture samples were collected from the patients in blood culture bottles (Becton and Dickinson, USA) and were processed in Bectec 9240, blood culture analyzer. The catheter tips and insertion site swabs were cultured on MacConky's and Blood Agar media and bacteria were identified by conventional microbiological method and API technique (API-System, La Balmes-Ies Grottes, France).
After the clinical diagnosis of septicemia was established, the patients were given antipyretics, amikacin and vancomycin, administered intravenous (i.v.) post-dialysis with the dosage adjusted by regular monitoring of plasma peak and trough levels. Patients with persistent hypotension or mental changes were hospitalized. Therapy with amikacin and vancomycin was continued post-HD for five successive dialysis sessions in patients having temporary vascular access (FC and SC) infections and for seven successive dialysis sessions in patients with permanent vascular access (PC, AVG and AVF) infections awaiting culture and sensitivity results. Therapy was appropriately modified once the culture and sensitivity reports were available. Peripheral blood cultures were repeated on the 45th day after completion of antibiotic therapy. A cure was defined as a 45 days symptomfree interval after antibiotic therapy was completed whereas any bacteremia caused by the original organism occurring within 45 days of treatment was considered a treatment failure.
Statistical analysis was performed using Web X2 Calculator, Georgetown University, Washington DC, USA.
| Results|| |
The present study included 209 patients on long-term HD whose mean age was 47.5 years (range 15-85 years). Of these, 85 (40.6%) patients had clinically suspected and bacteriologically proven septicemia and included 31 (36.2%) males and 54 (63.7%) females. Overall, 124 episodes of septicemia with an average of 1.23 per 100 patientmonths during the four-year (10032 patientmonths) study period, were recorded [Table - 1]. Male patients (31/96) had 45 episodes during 4608 patient-months recording 0.97 episodes per 100 patient-months and female patients (54/113) had 79 episodes of septicemia over 5,424 patient-months documenting 1.45 episodes per 100 patientmonths with relative risk (RR) of 1.48, 95% Confidence Interval (CI) of 1.045-2.096 and P-value < 0.02, [Table - 1].
The maximum numbers of episodes (54) of VARS were observed in the FC group (43.5%), followed by 35 (28.2%) in the SC group. Lowest number of episodes (11) was observed in the AVF group (8.8%). The number of episodes of VARS among the patients having PC fell between these two groups.
S. aureus was the commonest organism associated with septicemia in the SC and PC groups. Gram-negative bacilli were the commonest organisms responsible in the FC group while the organisms causing VARS in the AVG and AVF groups varied.
The maximum number of VARS episodes was caused by S. aureus (29.0%) in all the vascular access groups. The other common bacteria associated with VARS were Pseudomonas aeruginosa 5.3%), Acinetobactor (12.9%), Escherechia coli (10.4%) and S. epidermidis (9.6%) while Serratia marcesens was associated with one episode of septicemia [Table - 2].
Out of the total 124 episodes, 37 (29.8%) showed poor clinical response to antibiotic therapy and the access catheters had to be removed and the tips submitted for culture and sensitivity. All these 37 catheter tips were culture positive. S. aureus was grown from 18 (48.6%), P. aeruginosa and, S. epidermidis from 5 (13.5%) each and Enterobacter cloacae from 3 (8.1%) of the catheter tips. E. coli and Klebsiella pneumonae were each grown from two (5.4%) catheter tips while, Acinetobactor and Enterococcus faecalis were each grown from one (2.7%). Of the 37 episodes where catheter tip cultures were positive, 11 had the same organisms grown from both the catheter tips and swabs taken from the catheter insertion site.
The majority, 47 (37.9%) of the episodes of VARS, occurred among the patients over 50 years of age, while 37 (25.1%) were observed among patients having Type II diabetes mellitus.
Antibiotic access salvage was found to be successful with good outcome of preservation of vascular access sites in 48/85 (56.4%). However, a failure rate of 37/85 (43.6%) leading to loss of the accesses was also encountered.
A mortality rate of 25.9% (22/85) among the patients having VARS while being on antibiotics was observed in our study. S. aureus was found to be associated with 13/22 (59.1%) deaths while 9/22 (40.9%) deaths were recorded in the non-S. aureus group [RR-1.44, 95% CI(0.784-2.661)].
| Discussion|| |
Vascular access related septicemia is one of the leading causes of mortality and morbidity as well as the increase in the cost caused by hospitalizations, among ESRD patients on HD.  In the present study, 85 (40.6%) patients on HD had septicemia.
The average number of episodes among these patients was 1.23 per 100 patientmonths which is lower than the 3.6% reported earlier by Nsouli et al.  The episodes of VARS were found to be significantly higher (P<0.02) among female patients recording a 48% greater risk [RR1.48, 95% CI (1.045-2.096)] of developing VARS than the male patients (assigned RR of one). This could possibly be due to the prevailing local social, cultural and educational factors.
The high occurrence of VARS in the FC group caused predominantly by gramnegative bacilli is due to the close proximity to the perineal area contaminated with enteric group of organisms. In patients with AVF, 11 (8.8%) episodes of septicemia were observed. One of the patients had septicemia due to Hemophilus influenzae-B which resulted in severe respiratory tract infection due to this organism. Among the remaining 10 patients in this group, seven had UTI caused by the same organism that was isolated from the blood. Thus, only three patients from this group had septicemic episodes, truly related to vascular access which, could perhaps be due to repeated cannulations and cross infection of AVF during HD. Thus, the AVF group had the least number of septicemic episodes. This is in keeping with the recommendations by the DOQI guideline working panel as the most reliable and safe form of vascular access.
In the present study, S. aureus was responsible for 36 (29.1%) of the VARS episodes. S. aureus has been reported also by others to be commonest bacteria associated with VARS. , The next commonest group of organisms associated with VARS were gram-negative bacilli. Similar observations have been reported by other workers. ,
We started our patients with septicemia empirically on amikacin and vancomycin pending the culture and sensitivity report. This combination was considered appropriate considering the observation that S. aureus and gram negative bacilli were the commonest bacteria associated with VARS. The majority of the gram negative bacilli together with the Pseudomonas species isolated at this center were sensitive to aminoglycosides while resistance to third generation cephalosporins and quinolones was frequently observed. Amikacin was chosen due to its relatively long half-life enabling us to maintain effective therapeutic plasma drug levels between two dialysis sessions. Moreover, it is also valuable against penicillinase producing S. aureus and it has proven efficacy even against those gram-negative bacilli that are resistant to gentamycin (amikacin is resistant to eight of the nine classified aminoglycoside inactivating enzymes whereas gentamycin is inactivated by five). Staphylococcal septicemia is a potentially life-threatening condition leading to high mortality among all age-groups.  Since all the staphylococcal strains including methicillin-resistant S. aureus (MRSA) isolated at this center were sensitive to vancomycin, this was preferred over other anti-staphylococcal antibiotics in our study. Vancomycin has been reported to be the treatment of choice for staphylococcal septicemia among HD patients. 
Vascular access salvage through empirical amikacin-vancomycin therapy was successful in 48/85 (56.4%) patients having VARS, which is higher than the 30% success reported by Marr et al.  This maybe due to inclusion of AVF and AVG in our study, as these cases generally have low grade infection, with the consequent higher access salvage rates in comparison to vascular catheters. 
Mortality of 22/85 (25.9%) observed in this study was higher in patients with S. aureus associated VARS than those caused by non-S. aureus group [13/22 (59.1%) versus 9/22 (40.9%)]. Kirkland et al, have observed "Dialysis-access infection" related mortality of 20%, which is comparable to that reported in our study. 
In conclusion, AVFs are the safest of all vascular accesses associated with the lowest infection rates and therefore must be preferred and promoted as vascular access of choice. Temporary vascular catheters have the highest infection rates and thus, their use should be minimized. S. aureus was the commonest causative organism and was associated with a high mortality. An early empirical antibiotic therapy could effectively salvage vascular accesses in at least 50% of cases. Thus, an attempt at antibiotic access salvage may be recommended before actually sacrificing the vascular access in patients with VARS.
| References|| |
|1.||Schwab SJ. Vascular accesses for hemodialysis. Kidney Int 1999;55:2078-90. |
|2.||NKF-DOQI. Clinical practice guidelines for vascular access VIII , timing of access placement. Am J Kidney Dis 1997;30(Suppl 3):S160. |
|3.||Tanriover B, Carlton D, Saddekni S, et al. Bacteremia associated with tunneled dialysis catheters: comparison of two treatment strategies. Kidney Int 2000;57:2151-5. |
|4.||Powe NR, Jaar B, Furth SL, Hermann J, Briggs W. Septicemia in dialysis patients: incidence, risk factors and prognosis. Kidney Int 1999; 55:1081-90. |
|5.||Kirkland KB, Saxton DF. Dialysis access infection. In: Conlon PJ, Nicholson MI, Schwab S. Hemodialysis vascular access practice and problems. Oxford, New York: Oxford University Press 2000;pp 85-100. |
|6.||Kessler M, Hoen B, Mayeux D, Hestin D, Fontenaille C. Bacteremia in patients on chronic hemodialysis. A multi-center prospective survey. Nephron 1993;64:95-100. |
|7.||Dobkin JF, Miller MH, Steigbigel NH. Septicemia in patients on chronic hemodialysis. Ann Intern Med 1978;88:28-33. |
|8.||Nielsen J, Ladefoged SD, Kolmos HJ. Dialysis catheter-related septicemia-focus on Staphylococcus aureus septicemia. Nephrol Dial Transplant 1998;13:2847-52. |
|9.||Behrendt G, Scheider S, Brodt HR, Just-Nubling G, Shah PM. Influence of antimicrobial treatment on mortality in septicemia. J Chemother 1999; 11:179-86. |
|10.||Marr KA, Saxton DJ, Conlon PJ, et al. Catheter related bacteremia and outcome of attempted catheter salvage in patients undergoing hemodialysis. Ann Intern Med 1997;127:275-80. |
|11.||Raad II, Baba M, Bodey GP. Diagnosis of catheter-related infections: the role of surveillance and targeted quantitative skin cultures. Clin Infect Dis 1995;20:593-7. |
|12.||Rodriguez Guardado-A, Carton JA, Lopez PB, Casado L, Perez F, Aguado S. Bacteremia in patients undergoing chronic hemodialysis in a 16 year period. Rev Clin Esp 1997;197:484-9. |
|13.||Nsouli KA, Lazarus M, Schoenbaum SC, Gottieb MN, Lowrie EG, Shocair M. Bacteremic infection in hemodialysis. Arch Intern Med 1979;139:1255-8. |
|14.||Peacock SJ, Curtis N, Berendt AR, Bowler IC, Winearls CG, Maxwell P. Outcome following hemodialysis catheter-related Staphylococcus aureus bacteremia. J Hosp Infect 1999;41:223-8. |
|15.||Raad II, Sabbagh MF. Optimal duration of therapy for catheter-related Staphylococcus aureus bacteremia: a study of 55 cases and review. Clin Infect Dis 1992;14:75-82. |
|16.||Beck-Sague CM, Jarvis WR, Bland LA, Arduino MJ, Aguero SM, Verosic G. Outbreak of gram negative bacteremia and pyrogenic reactions in a hemodialysis center. Am J Nephrol 1990;10:397-403. |
|17.||Green K, Schulman G, Haas DW, Schaffner W, D'Agata EM. Vancomycin prescribing practices in hospitalized chronic hemodialysis patients. Am J Kidney Dis 2000;35:64-8. |
|18.||Churchill DN, Taylor DW, Cook RJ, et al. Canadian hemodialysis morbidity study. Am J Kidney Dis 1992;19:214-34. |
Anil K Saxena
Division of Nephrology, King Fahad Hospital, Al-Hasa 31982, Hofuf
[Table - 1], [Table - 2]